Victron Smart “Shuntapede” test results

This is a short form version of the results, good for those that dislike long posts.

TLDR… Victron Shunts are good and can be trusted, to a point.

We set up almost 30 Victron Smart Shunts in series to check their current accuracy against each other. It turns out they are very accurate, however they didn’t quite meet the specification quoted by Victron. We achieved results close to about 1.5% variation across the range. The specification (+-0.4%) suggests 0.8% difference should be the maximum we should see.

Everything possible was done to remove any bias, including resetting the Zero Current for each unit in-situ. This was done prior to all tests and whenever power was removed from the shunts. We noticed some slight changes in the Zero Current when power was removed and restored, however it was not always consistent.

We don’t believe there’s any interference of the internal power supplies of each shunt on another shunts readings.

There was no noticeable variation in temperature between the shunts throughout the testing. So all shunts increased temperature at the same rate, negating any influence.

Testing was also done at various current settings to see if there was any difference. Generally it was done at approx. 60A, 80A and 120A depending on the specific test setup and equipment used. There was no noticeable difference due to test current.

The shunts had positions swapped to check for any positional bias. There was no noticeable difference related to the position in the string.

Different power sources were tested to eliminate source noise being an issue, these included a switch mode power supply, 2 x switch mode power supplies in parallel, and a battery. No noticeable difference was observed.

Different loads were also tested to eliminate load noise being an issue. This included an inverter running a fan heater, an inverter running other loads, and a DC-DC converter charging a battery. There were no observed differences based on load type.

Once the first test was completed, we purchased extra shunts and tested those, then chose the closest 24 shunts, which then all tested within 1% of each other. These will form the test bench shunts and there will be no need for any post test corrections due to their consistency.

Voltage reading differences of up to 0.03v have been observed between shunts @ approx. 14.50v. This was observed in a separate test.

This ultimately means that the same battery may test between 308Ah and 311Ah (example only) depending on the specific shunt being used. While it’s not perfect, it’s more than adequate for what we require here, and other influences such as age, temperature and load current will have a bigger effect.

So we’re going ahead with using these from this point on. All batteries produced from June 2024 onwards will have some slightly more consistent capacity ratings that can also be achieved by all customers.

We’re also going to look to see if fitting Victron Orion XS 50A units will be possible instead of the existing Renogy DC-DC. It would need a circuit revision as the system uses a common positive as it is now, we may need to make it common negative (TBC). If this works out we can prevent even more losses and remove somewhat stressed Renogy equipment. It should also increase system efficiency from about 70% up to almost 90% (hoped) so less energy will need to be used and less heat produced.